Abstract:Non-linear convection-reaction-diffusion (CRD) partial differential equations (PDEs) are crucial for modeling complex phenomena in fields such as biology, ecology, population dynamics, physics, and engineering. Numerical approximation of these non-linear systems is essential due to the challenges of obtaining exact solutions. Traditionally, the Galerkin finite element method (GFEM) has been the standard computational tool for solving these PDEs. With the advancements in machine learning, Physics-Informed Neural Network (PINN) has emerged as a promising alternative for approximating non-linear PDEs.
In this study, we compare the performance of PINN and GFEM by solving four distinct one-dimensional CRD problems with varying initial and boundary conditions and evaluate the performance of PINN over GFEM. This evaluation metrics includes error estimates, and visual representations of the solutions, supported by statistical methods such as the root mean squared error (RMSE), the standard deviation of error, the the Wilcoxon Signed-Rank Test and the coefficient of variation (CV) test.
Our findings reveal that while both methods achieve solutions close to the analytical results, PINN demonstrate superior accuracy and efficiency. PINN achieved significantly lower RMSE values and smaller standard deviations for Burgers' equation, Fisher's equation, and Newell-Whitehead-Segel equation, indicating higher accuracy and greater consistency. While GFEM shows slightly better accuracy for the Burgers-Huxley equation, its performance was less consistent over time. In contrast, PINN exhibit more reliable and robust performance, highlighting their potential as a cutting-edge approach for solving non-linear PDEs.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to evaluate the performance differences between Physics - Informed Neural Networks (PINNs) and Galerkin Finite Element Method (GFEM) in solving nonlinear convection - reaction - diffusion (CRD) partial differential equations (PDEs). Specifically, by solving four different one - dimensional CRD problems with different initial and boundary conditions, the researchers compared the performance of these two methods in error estimation, visual representation, and statistical methods (such as root - mean - square error (RMSE), standard deviation of error, Wilcoxon signed - rank test, and coefficient of variation (CV) test).
### Research Background
Nonlinear convection - reaction - diffusion partial differential equations are used to model complex phenomena in fields such as biology, ecology, population dynamics, physics, and engineering. Due to the challenges of obtaining exact solutions, numerical approximation of these nonlinear systems has become crucial. Traditionally, Galerkin Finite Element Method (GFEM) is the standard computational tool for solving these PDEs. With the progress of machine learning, Physics - Informed Neural Networks (PINNs) have become a promising alternative for approximating nonlinear PDEs.
### Research Objectives
The main objective of this paper is to conduct a systematic performance, accuracy, and consistency evaluation of PINNs and GFEM to determine their performance in solving a series of nonlinear PDEs. Specifically, the researchers selected the following four one - dimensional CRD problems for comparison:
1. **Burgers’ Equation**: Describes the propagation of waves and interactions in nonlinear media.
2. **Fisher’s Equation**: Widely used in heat conduction, mass transfer, biology, and ecology, especially for simulating the spread of favorable gene traits in populations.
3. **Burgers - Huxley Equation**: Combines the characteristics of Burgers and Huxley equations and is applicable in fields such as fluid mechanics, chloride diffusion in concrete, mathematical biology, air pollutant emissions, population genetics, nonlinear acoustics, and combustion theory.
4. **Newell - Whitehead - Segel Equation**: Used to study the interaction between diffusion and nonlinear reaction terms and provides valuable insights in fluid mechanics, biological systems, and financial modeling.
### Main Findings
The research results show that although both methods can approximate the analytical results, PINNs exhibit higher accuracy and efficiency. Specifically:
- PINNs achieve significantly lower RMSE values and smaller standard deviations in Burgers’ Equation, Fisher’s Equation, and Newell - Whitehead - Segel Equation, indicating higher accuracy and better consistency.
- Although GFEM shows slightly higher accuracy in Burgers - Huxley Equation, its performance has poorer stability over time.
- PINNs demonstrate more reliable and robust performance, highlighting their potential as a cutting - edge method for solving nonlinear PDEs.
### Conclusion
Through a multi - method evaluation framework, the researchers provided detailed numerical comparisons, graphical visualizations, and strict statistical analyses to verify the performance of the two methods. The results show that PINNs have significant advantages in solving nonlinear PDEs, especially in complex geometries and high - dimensional spaces.